CN110084104A - The method for carrying out the detection of naval vessel moving-target based on gaussian pyramid light stream - Google Patents
The method for carrying out the detection of naval vessel moving-target based on gaussian pyramid light stream Download PDFInfo
- Publication number
- CN110084104A CN110084104A CN201910197490.1A CN201910197490A CN110084104A CN 110084104 A CN110084104 A CN 110084104A CN 201910197490 A CN201910197490 A CN 201910197490A CN 110084104 A CN110084104 A CN 110084104A
- Authority
- CN
- China
- Prior art keywords
- doubtful
- phase
- target
- light stream
- ship target
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/13—Satellite images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/46—Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/07—Target detection
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Astronomy & Astrophysics (AREA)
- Remote Sensing (AREA)
- Image Analysis (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The embodiment of the invention provides a kind of methods for carrying out the detection of naval vessel moving-target based on gaussian pyramid light stream, it include: the optics two field pictures for obtaining satellite acquisition, the extra large land of two field pictures progress is divided and shields land area, two field pictures are respectively phase 1 and phase 2, the extraction of doubtful Ship Target candidate regions is carried out to the image after the segmentation of extra large land with constant false alarm rate algorithm, obtain the position of doubtful Ship Target, it is right, the doubtful Ship Target extracted carries out optical flow tracking, according to light stream, the position of Ship Target doubtful in phase 1 is carried out to the position mark of doubtful Ship Target in phase 2, reject the part for exceeding 2 boundary of phase in mark position, according to the direction of the doubtful Ship Target of optical flow computation, after removing the strong pixel of orientation consistency, as naval vessel moving-target.The embodiment of the present invention overcomes traditional ShipTargets detection and tracking technology and is difficult to reach the defect of ideal effect, completes the quick detection to naval vessel moving-target.
Description
Technical field
The present invention relates to pelagic region wide format images detection technique fields, are based on gaussian pyramid light more particularly to one kind
The method that stream carries out the detection of naval vessel moving-target.
Background technique
Satellite imagery mainly uses optical imagery and synthetic aperture radar (SAR) imaging technique from space to the earth or the moon
Equal celestial bodies are imaged, for remote sensing, environmental monitoring or military surveillance.Satellite is from top to bottom to pelagic region Ship Target
When imaging, the complicated cloud that is presented and Ship Target and the low-resolution image frame sequence deposited, in candidate region Ship Target with
The feature difference of false-alarm is unobvious, and the object filtering technology based on characteristics of image is difficult to reach ideal the selection result.Multiple
In the image sequence of miscellaneous cloud background, candidate regions are carried out according to otherness existing for the mass motion direction of moving target and cloud background
The screening operation in domain.
Currently, common moving-target discovery technique is all based on the complete image detecting technique of traversal to realize.However
Not only operand is big for this method, also relatively high to the resource allocation request of hardware, in the case where the preferable background interference of environment is small
It was found that effect it is preferable, still, easily erroneous detection and missing inspection when there is the interference of big cloudlet in background.
Therefore, how innovatively a technical problem that needs to be urgently solved by technical personnel in the field at present is exactly:
A kind of method for proposing effective naval vessel moving-target detection is met in practical application with overcoming defect of the existing technology
Greater demand.
Summary of the invention
In view of the above problems, it proposes the embodiment of the present invention and overcomes the above problem or at least partly in order to provide one kind
The method for carrying out the detection of naval vessel moving-target based on gaussian pyramid light stream to solve the above problems, the embodiment of the present invention effectively overcome
High rail satellite imaging data resolution ratio is low, and visual field is wide and scene is complicated and changeable, traditional marine vessel moving-target detection and
Tracking technique is difficult to the problem of reaching ideal effect.
To solve the above-mentioned problems, the invention discloses one kind carries out the detection of naval vessel moving-target based on gaussian pyramid light stream
Method, comprising:
S101, the optics two field pictures for obtaining satellite acquisition divide the extra large land of two field pictures progress and shield land area；
The two field pictures are respectively phase 1 and phase 2；
S102, it is directed to phase 1, the extraction of doubtful Ship Target is carried out with constant false alarm rate algorithm；
S103, optical flow tracking is carried out to the doubtful Ship Target that S102 is extracted；
S104, according to S103 optical flow tracking as a result, position of the Ship Target doubtful in phase 1 in phase 2 is carried out
Label；
S105, rejecting S104 mark the part for exceeding 2 boundary of phase in the position of doubtful Ship Target；
S106, according to S103 optical flow tracking as a result, calculate phase 2 in doubtful Ship Target direction, remove direction one
After the strong pixel of cause property, retaining is naval vessel moving-target.
Preferably, the doubtful Ship Target extracted to S103 carries out optical flow tracking, comprising:
Original image establishes pyramid figure layer as the 0th layer, using recursive form；
Initializing the top light stream estimated value of pyramid is 0, and top calculated result passes to down as initial value
One tomographic image, according to the light stream of this layer of calculation of initial value, then light stream that this layer is obtained is as next layer of initial value, until most
Low layer original image layer, all layers of segmentation light stream superposition value is as last light stream.
Preferably, the extraction that doubtful Ship Target candidate regions are carried out to image obtained by S102, extracted doubtful warship
Ship target candidate area is the region of the big Yu Haiyang's gray scale of gray scale.
Preferably, the implementation of the shielding land area are as follows: land is filled with ocean.
Preferably, the optics two field pictures are continuous continual two frame.
Preferably, the image after the segmentation to the sea S102 land carries out the extractions of doubtful Ship Target candidate regions, using mentioning
The mode of connected domain is taken to complete.
Preferably, the connected domain is four connected region or eight connected region.
The embodiment of the present invention includes following advantages:
The present invention carries out extra large land segmentation to image by will acquire image, with Hai Luku, in extra large land on the basis of segmentation according to
CFAR algorithm extracts candidate regions, and carrying out optical flow tracking to obtained suspected target must in order to quickly find the position of Ship Target
Cloud must be interfered and be removed, for this purpose, the direction of all suspected targets in image and statistical analysis are calculated, because of cloud under big visual field
The direction of motion be it is the same, the target moved in big visual field only has cloud and naval vessel, and counting direction and rejecting cloud quickly to obtain
The position of Ship Target is obtained, realizes the quick detection to naval vessel moving-target.
Detailed description of the invention
Fig. 1 is a kind of embodiment one of method that moving-target detection in naval vessel is carried out based on gaussian pyramid light stream of the invention
Step flow chart；
Fig. 2 is a kind of embodiment two of method that moving-target detection in naval vessel is carried out based on gaussian pyramid light stream of the invention
Flow diagram；
Fig. 3 is a kind of embodiment two of method that moving-target detection in naval vessel is carried out based on gaussian pyramid light stream of the invention
The schematic diagram of the image instance of the adjacent remote sensing satellite of two frames of the acquisition；
Fig. 4 is a kind of embodiment two of method that moving-target detection in naval vessel is carried out based on gaussian pyramid light stream of the invention
The CFAR model schematic；
Fig. 5 is a kind of embodiment two of method that moving-target detection in naval vessel is carried out based on gaussian pyramid light stream of the invention
Described marks the image effect schematic diagram for paying close attention to Ship Target；
Fig. 6 is a kind of embodiment two of method that moving-target detection in naval vessel is carried out based on gaussian pyramid light stream of the invention
The optical flow tracking flow diagram.
Specific embodiment
In order to make the foregoing objectives, features and advantages of the present invention clearer and more comprehensible, with reference to the accompanying drawing and specific real
Applying mode, the present invention is described in further detail.
Core of the invention thought is to disclose a kind of method for carrying out the detection of naval vessel moving-target based on gaussian pyramid light stream,
This method will acquire image, carry out extra large land segmentation to image with Hai Luku；It is extracted on the basis of extra large land is divided according to CFAR algorithm
Candidate regions；Optical flow tracking is carried out to obtained suspected target；In order to quickly find the position of Ship Target, it is necessary to go cloud interference
It removes, for this purpose, the direction of all suspected targets in image and statistical analysis are calculated, because the direction of motion of cloud is one under big visual field
Sample, the target moved in big visual field only has cloud and naval vessel, and Ship Target can be quickly obtained by counting direction and rejecting cloud
Position.This method can effectively detect the position of Ship Target in wide format images, export the coordinate information of target.
Embodiment one
Referring to Fig. 1, a kind of method that moving-target detection in naval vessel is carried out based on gaussian pyramid light stream of the invention is shown
Embodiment one step flow chart, can specifically include following steps:
Step 101, the optics two field pictures for obtaining satellite acquisition divide the extra large land of two field pictures progress and shield land area
Domain；The two field pictures are respectively phase 1 and phase 2, resolution ratio per inch pixel number (Dots Per Inch, abbreviation DPI)
200 or more, more than 1080 × 720 pixel of size；
Step 102, for phase 1, the extraction of doubtful Ship Target is carried out with constant false alarm rate algorithm；
The extraction that doubtful Ship Target candidate regions are carried out to image obtained by S102, extracted doubtful Ship Target are waited
Constituency is the region of the big Yu Haiyang's gray scale of gray scale, the implementation of the shielding land area are as follows: filled out with ocean to land
It fills.
Two frame optical imagerys targeted in the present embodiment are continuous continual two frame simultaneously.
Image after the segmentation to the sea S102 land carries out the extraction of doubtful Ship Target candidate regions, using extraction connected domain
Mode complete, the connected domain be four connected region or eight connected region.
Step S103 carries out optical flow tracking to the doubtful Ship Target that S102 is extracted；
Step S104, according to S103 optical flow tracking as a result, position by Ship Target doubtful in phase 1 in phase 2
It is marked；
Step S105 rejects S104 and marks the part for exceeding 2 boundary of phase in the position of doubtful Ship Target；
Step S106, according to S103 optical flow tracking as a result, calculating the direction of doubtful Ship Target in phase 2, removal side
After the pixel strong to consistency, naval vessel moving-target is used as by what is retained.
In practical operation, the feature indexed by the direction of light stream and cloud, the light stream orientation consistency of doubtful Ship Target is strong
Pixel is cloud index, and the index for removing cloud can quickly find the position of dynamic Ship Target.
Embodiment two
Referring to fig. 2, a kind of method that moving-target detection in naval vessel is carried out based on gaussian pyramid light stream of the invention is shown
Embodiment two flow diagram, concrete processing procedure is as follows:
S1: video of the power line containing foreign matter is acquired using camera, acquires the resolution ratio of every frame image of video at least
More than 200dpi, 1080 × 720 pixel of size, Fig. 3 is the image instance of the adjacent remote sensing satellite of two frames of acquisition, to 1 He of phase
2 two field pictures of phase carry out extra large land using digital elevation model (Digital Elevation Model, vehicle economy M) and divide, screen
Cover land area；When actual treatment sea land is divided, the value of ocean is 0；The value on land is 1.
S2: the phase 1 in image obtained to S1 obtains doubtful Ship Target candidate regions using CFAR algorithm；Gray value
Big Yu Haiyang's gray value is used as Ship Target candidate regions, then extracts doubtful Ship Target；
S21: assuming that the target area of Ship Target candidate regions, protection zone and background area are denoted as Target respectively
Area, Protect area and Background area.It is carried out as shown in figure 4, sea clutter statistics chooses Gaussian distribution model
CFAR detection, will obtain different distributed models to different candidate targets, will also obtain different CFAR threshold values, CFAR
Thresholding has adaptive characteristic.
Two-parameter CFAR detector formula are as follows:
Wherein xtIt is pixel to be tested, μbIt is background mean value, σbIt is that background standard is poor, t is detector design parameter.CFAR
Model is as shown in Figure 4.
S22: after obtaining candidate regions, connected domain is extracted, position and the target number of doubtful Ship Target are obtained.In,
Connected domain is eight connected region or four connected region.
S3: the doubtful Ship Target obtained after S2 is handled carries out optical flow tracking, and flow chart is as shown in Figure 6；
S31: pyramid is established；
Original image enables I as the 0th layer0=I is the 0th layer of image, and image width is defined as W0=W, height are defined as H0=
H establishes pyramid, I using recursive formLIt is IL-1Layer is derived.Derivation formula is as follows:
In specific processing, 4 layers are typically set up, it may be assumed that I0、I1、I2And I3Layer.
S32: pyramid tracking:
Light stream is calculated on minimum tomographic image, passes to next layer of figure for upper one layer of calculated result as initial value
Picture, according to the light stream of this layer of calculation of initial value, then light stream that this layer is obtained is as next layer of initial value, until minimum one layer
(original image layer), the light stream that this layer is calculated are last light stream.
1) light stream estimator is initialized
Initializing the light stream estimated value in top layer images is 0, and pyramid gradually becomes smaller from bottom to high-rise size, reduces
Light stream value, therefore top light stream estimated value is set as 0.
Define velocity vector
In neighborhood the matching error of all pixels point and are as follows:
Wherein A and B is the gray value at coordinate.
Optical flow computation to every layer is exactly that the derivative of matching error sum in neighborhood is sought using least square method, in optimal solution position
Setting derivative is 0, similarity highest between matching error minimum, that is, two field pictures corresponding points.
The derivative for finding out error sum, is unfolded using Taylor's formula, obtains the optimal solution of light stream vectors:
Definition
The light stream of every tomographic image is arrived according to what above formula calculated.δ I is the difference of two frame same layer gray value of images, IxAnd Iy
It is gradient component of the image in the point respectively.
2) feedback arrives next layer
The estimated value of next layer of light stream is set.
gL-1=2 (gL+dL)
Final light stream is exactly all layers of segmentation light stream superposition value.
3) relationship of light stream and pixel
Light stream is exactly the x of each pixel on image during picture moving, y displacement amount.Therefore according to above-mentioned steps
Corresponding object pixel in next frame can be obtained by obtaining light stream.
S4: according to S3 optical flow tracking as a result, position of the Ship Target doubtful in phase 1 in phase 2 is marked,
As shown in Figure 3；
S5: the position in phase 2 that the suspected target in phase 1 is obtained according to light stream sees if fall out phase 2
Boundary is rejected if exceeding.
S6: according to S3 optical flow tracking as a result, the direction of doubtful Ship Target in phase 2 can be calculated, suspected target is counted
Direction, after the strong pixel of removal orientation consistency, the Ship Target paid close attention to can be marked, as shown in Figure 5；It is abundant
The efficiency and applicability for verifying the method for the present invention, the method for the present invention is tested, the side of the present invention it can be seen from experimental result
Method can preferably detect Ship Target in wide cut scene.
It should be noted that for simple description, therefore, it is stated as a series of action groups for embodiment of the method
It closes, but those skilled in the art should understand that, embodiment of that present invention are not limited by the describe sequence of actions, because according to
According to the embodiment of the present invention, some steps may be performed in other sequences or simultaneously.Secondly, those skilled in the art also should
Know, the embodiments described in the specification are all preferred embodiments, and the related movement not necessarily present invention is implemented
Necessary to example.
All the embodiments in this specification are described in a progressive manner, the highlights of each of the examples are with
The difference of other embodiments, the same or similar parts between the embodiments can be referred to each other.
Those skilled in the art after considering the specification and implementing the invention disclosed here, will readily occur to of the invention its
Its embodiment.The present invention is directed to cover any variations, uses, or adaptations of the invention, these modifications, purposes or
Person's adaptive change follows general principle of the invention and including the undocumented common knowledge in the art of the disclosure
Or conventional techniques.The description and examples are only to be considered as illustrative, and true scope and spirit of the invention are by following
Claim is pointed out.
It should be understood that the present invention is not limited to the precise structure already described above and shown in the accompanying drawings, and
And various modifications and changes may be made without departing from the scope thereof.The scope of the present invention is limited only by the attached claims
The foregoing is merely presently preferred embodiments of the present invention, is not intended to limit the invention, it is all in spirit of the invention and
Within principle, any modification, equivalent replacement, improvement and so on be should all be included in the protection scope of the present invention.
Above to it is provided by the present invention it is a kind of based on gaussian pyramid light stream carry out the detection of naval vessel moving-target method, into
It has gone and has been discussed in detail, used herein a specific example illustrates the principle and implementation of the invention, the above implementation
The explanation of example is merely used to help understand method and its core concept of the invention；Meanwhile for the general technology people of this field
Member, according to the thought of the present invention, there will be changes in the specific implementation manner and application range, in conclusion this explanation
Book content should not be construed as limiting the invention.
Claims (7)
1. a kind of method for carrying out the detection of naval vessel moving-target based on gaussian pyramid light stream characterized by comprising
S101, the optics two field pictures for obtaining satellite acquisition divide the extra large land of two field pictures progress and shield land area；It is described
Two field pictures are respectively phase 1 and phase 2；
S102, it is directed to phase 1, the extraction of doubtful Ship Target is carried out with constant false alarm rate algorithm；
S103, optical flow tracking is carried out to the doubtful Ship Target that S102 is extracted；
S104, according to S103 optical flow tracking as a result, position of the Ship Target doubtful in phase 1 in phase 2 is marked；
S105, rejecting S104 mark the part for exceeding 2 boundary of phase in the position of doubtful Ship Target；
S106, according to S103 optical flow tracking as a result, calculate phase 2 in doubtful Ship Target direction, remove orientation consistency
After strong pixel, retaining is naval vessel moving-target.
2. the method according to claim 1, wherein the doubtful Ship Target extracted to S103 carries out light stream
Tracking, comprising:
Original image establishes pyramid figure layer as the 0th layer, using recursive form；
Initializing the top light stream estimated value of pyramid is 0, and top calculated result passes to next layer as initial value
Image, according to the light stream of this layer of calculation of initial value, then light stream that this layer is obtained is as next layer of initial value, until minimum one
Layer original image layer, all layers of segmentation light stream superposition value is as last light stream.
3. method according to claim 1 or 2, which is characterized in that described to carry out doubtful naval vessel mesh to image obtained by S102
The extraction of candidate regions is marked, extracted doubtful Ship Target candidate regions are the region of the big Yu Haiyang's gray scale of gray scale.
4. according to the method described in claim 3, it is characterized in that, the implementation of the shielding land area are as follows: use ocean
Land is filled.
5. according to the method described in claim 3, it is characterized in that, the optics two field pictures are continuous continual two frame.
6. according to the method described in claim 3, it is characterized in that, the image after the segmentation to the sea S102 land carries out doubtful warship
The extraction in ship target candidate area is completed by the way of extracting connected domain.
7. according to the method described in claim 6, it is characterized in that, the connected domain is four connected region or eight connected region.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910197490.1A CN110084104A (en) | 2019-03-15 | 2019-03-15 | The method for carrying out the detection of naval vessel moving-target based on gaussian pyramid light stream |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910197490.1A CN110084104A (en) | 2019-03-15 | 2019-03-15 | The method for carrying out the detection of naval vessel moving-target based on gaussian pyramid light stream |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110084104A true CN110084104A (en) | 2019-08-02 |
Family
ID=67412594
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910197490.1A Pending CN110084104A (en) | 2019-03-15 | 2019-03-15 | The method for carrying out the detection of naval vessel moving-target based on gaussian pyramid light stream |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110084104A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111220956A (en) * | 2019-11-08 | 2020-06-02 | 北京理工雷科电子信息技术有限公司 | Method for removing sea detection land target by airborne radar based on geographic information |
| CN111899285A (en) * | 2020-07-08 | 2020-11-06 | 浙江大华技术股份有限公司 | Method and device for determining tracking track of target object and storage medium |
| CN111973919A (en) * | 2020-06-24 | 2020-11-24 | 江苏深渡消防装备科技有限公司 | Fire source accurate positioning method based on image recognition |
| CN112257505A (en) * | 2020-09-17 | 2021-01-22 | 昆明物理研究所 | False target identification method applied to infrared panoramic system and storage medium |
| CN113112481A (en) * | 2021-04-16 | 2021-07-13 | 北京理工雷科电子信息技术有限公司 | Mixed heterogeneous on-chip architecture based on matrix network |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005182375A (en) * | 2003-12-18 | 2005-07-07 | Nissan Motor Co Ltd | Driver expression recognizing device |
| JP4111080B2 (en) * | 2003-06-27 | 2008-07-02 | 日産自動車株式会社 | Armpit state detection device |
| CN101520896A (en) * | 2009-03-30 | 2009-09-02 | 中国电子科技集团公司第十研究所 | Method for automatically detecting cloud interfering naval vessel target by optical remote sensing image |
| CN102200575A (en) * | 2010-12-02 | 2011-09-28 | 南京大学 | Image ship detection method based on constant false alarm rate |
| CN102968798A (en) * | 2012-12-12 | 2013-03-13 | 北京航空航天大学 | SAR (Synthetic Aperture Radar) image sea-land segmentation method based on wavelet transform and OTSU threshold |
| US20150036888A1 (en) * | 2013-07-31 | 2015-02-05 | Trimble Navigation Ltd. | Sequential rolling bundle adjustment |
| CN106845364A (en) * | 2016-12-28 | 2017-06-13 | 中国航天电子技术研究院 | A kind of fast automatic object detection method |
| CN106875420A (en) * | 2016-12-29 | 2017-06-20 | 北京理工雷科电子信息技术有限公司 | A kind of aerial moving-target sequence image emulation mode based on remote sensing platform |
| CN108257149A (en) * | 2017-12-25 | 2018-07-06 | 翟玉婷 | A kind of Ship Target real-time tracking detection method based on optical flow field |
| CN108805904A (en) * | 2018-05-25 | 2018-11-13 | 中国空间技术研究院 | A kind of moving ship detection and tracking based on satellite sequence image |
| CN109117802A (en) * | 2018-08-21 | 2019-01-01 | 东北大学 | Ship Detection towards large scene high score remote sensing image |
-
2019
- 2019-03-15 CN CN201910197490.1A patent/CN110084104A/en active Pending
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4111080B2 (en) * | 2003-06-27 | 2008-07-02 | 日産自動車株式会社 | Armpit state detection device |
| JP2005182375A (en) * | 2003-12-18 | 2005-07-07 | Nissan Motor Co Ltd | Driver expression recognizing device |
| CN101520896A (en) * | 2009-03-30 | 2009-09-02 | 中国电子科技集团公司第十研究所 | Method for automatically detecting cloud interfering naval vessel target by optical remote sensing image |
| CN102200575A (en) * | 2010-12-02 | 2011-09-28 | 南京大学 | Image ship detection method based on constant false alarm rate |
| CN102968798A (en) * | 2012-12-12 | 2013-03-13 | 北京航空航天大学 | SAR (Synthetic Aperture Radar) image sea-land segmentation method based on wavelet transform and OTSU threshold |
| US20150036888A1 (en) * | 2013-07-31 | 2015-02-05 | Trimble Navigation Ltd. | Sequential rolling bundle adjustment |
| CN106845364A (en) * | 2016-12-28 | 2017-06-13 | 中国航天电子技术研究院 | A kind of fast automatic object detection method |
| CN106875420A (en) * | 2016-12-29 | 2017-06-20 | 北京理工雷科电子信息技术有限公司 | A kind of aerial moving-target sequence image emulation mode based on remote sensing platform |
| CN108257149A (en) * | 2017-12-25 | 2018-07-06 | 翟玉婷 | A kind of Ship Target real-time tracking detection method based on optical flow field |
| CN108805904A (en) * | 2018-05-25 | 2018-11-13 | 中国空间技术研究院 | A kind of moving ship detection and tracking based on satellite sequence image |
| CN109117802A (en) * | 2018-08-21 | 2019-01-01 | 东北大学 | Ship Detection towards large scene high score remote sensing image |
Non-Patent Citations (6)
| Title |
|---|
| BOUGUET J Y. 等: "Pyramidal Implementation of the affine Lucas Kanade Feature Tracker Description of the algorithm", 《INTEL CORPORATION》 * |
| HAICHAO LI 等: "Moving ship detection based on visual saliency for video satellite", 《2016 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM (IGARSS)》 * |
| ZBXZC: "目标跟踪之Lukas-Kanade光流法", 《HTTPS://BLOG.CSDN.NET/U014568921/ARTICLE/DETAILS/46638557?》 * |
| 刘思彤 等: "多特征融合的高分辨率遥感图像海陆分离", 《电光与控制》 * |
| 叶玲利: "内河视频监控中运动船舶检测和遮挡分离算法研究", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
| 吴琦颖 等: "一种新颖的海上运动目标实时检测方法", 《计算机工程与应用》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111220956A (en) * | 2019-11-08 | 2020-06-02 | 北京理工雷科电子信息技术有限公司 | Method for removing sea detection land target by airborne radar based on geographic information |
| CN111220956B (en) * | 2019-11-08 | 2021-11-09 | 北京理工雷科电子信息技术有限公司 | Method for removing sea detection land target by airborne radar based on geographic information |
| CN111973919A (en) * | 2020-06-24 | 2020-11-24 | 江苏深渡消防装备科技有限公司 | Fire source accurate positioning method based on image recognition |
| CN111899285A (en) * | 2020-07-08 | 2020-11-06 | 浙江大华技术股份有限公司 | Method and device for determining tracking track of target object and storage medium |
| CN111899285B (en) * | 2020-07-08 | 2023-03-14 | 浙江大华技术股份有限公司 | Method and device for determining tracking track of target object and storage medium |
| CN112257505A (en) * | 2020-09-17 | 2021-01-22 | 昆明物理研究所 | False target identification method applied to infrared panoramic system and storage medium |
| CN112257505B (en) * | 2020-09-17 | 2023-07-04 | 昆明物理研究所 | False target identification method and storage medium applied to infrared panoramic system |
| CN113112481A (en) * | 2021-04-16 | 2021-07-13 | 北京理工雷科电子信息技术有限公司 | Mixed heterogeneous on-chip architecture based on matrix network |
| CN113112481B (en) * | 2021-04-16 | 2023-11-17 | 北京理工雷科电子信息技术有限公司 | Hybrid heterogeneous on-chip architecture based on matrix network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110084104A (en) | The method for carrying out the detection of naval vessel moving-target based on gaussian pyramid light stream | |
| Xuan et al. | An improved canny edge detection algorithm | |
| Zhang et al. | A categorization of multiscale-decomposition-based image fusion schemes with a performance study for a digital camera application | |
| Chen et al. | Ship detection from coastal surveillance videos via an ensemble Canny-Gaussian-morphology framework | |
| CN108876723B (en) | Method for constructing color background of gray target image | |
| CN108981672A (en) | Hatch door real-time location method based on monocular robot in conjunction with distance measuring sensor | |
| CN109767454B (en) | Unmanned aerial vehicle aerial video moving target detection method based on time-space-frequency significance | |
| CN103927751A (en) | Water surface optical visual image target area detection method based on gradient information fusion | |
| CN108229342B (en) | Automatic sea surface ship target detection method | |
| CA2780595A1 (en) | Method and multi-scale attention system for spatiotemporal change determination and object detection | |
| CN110503620B (en) | Image fusion method based on Fourier spectrum extraction | |
| CN108109163A (en) | A kind of moving target detecting method for video of taking photo by plane | |
| WO2018127539A1 (en) | Methods and devices for extracting an object from a video sequence | |
| Zhao et al. | SAR ship detection based on end-to-end morphological feature pyramid network | |
| CN107862262A (en) | A kind of quick visible images Ship Detection suitable for high altitude surveillance | |
| Ding et al. | Coastline detection in SAR images using multiscale normalized cut segmentation | |
| CN106997583A (en) | A kind of underwater robot adaptive image enhancement and feature extracting method | |
| Li et al. | An oriented SAR ship detector with mixed convolution channel attention module and geometric nonmaximum suppression | |
| CN115760790A (en) | Airport pavement foreign matter detection method, server and storage medium | |
| CN110660022A (en) | Image super-resolution reconstruction method based on surface fitting | |
| CN109918968A (en) | A kind of Ship Target Detection method | |
| Prakash et al. | Vehicle license plate detection and recognition using non-blind image de-blurring algorithm | |
| Tian et al. | High confidence detection for moving target in aerial video | |
| CN109685825A (en) | Local auto-adaptive feature extracting method, system and storage medium for thermal infrared target tracking | |
| Xiang et al. | A modified joint trilateral filter based depth map refinement method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190802 |